High-precision Depth Research Articles

Dual Gm-APD Polarization Lidar to Acquire the Depth Image of Shallow Semitransparent Media With a Wide Laser Pulse

A polarization lidar with dual Geiger-mode avalanche photodiodes (Gm-APDs) is designed to obtain the high-precision depth image of shallow semitransparent media by using a wide laser pulse. The first surface of the semitransparent media is smooth and can nearly preserve the incident polarization, whereas the second surface is rough and can depolarize the incident polarization. A polarization splitting prism is utilized to separate the two echo signals, which will be detected by dual Gm-APDs. It gives the ability to obtain the extremely shallow depth independent of laser pulse width and dead time of Gm-APD. The theory of laser polarization transmission of dual Gm-APD polarization lidar is analyzed. Signal restoration & center-of-mass algorithm method is used to restrain the range walk error and obtain high range precision. The depth image of shallow semitransparent media with a thickness and distance of 10 cm and 6 m, respectively, is obtained in the experiment with a range precision of 1.1 cm. An experiment of real shallow water layer with different depth features is also performed to verify the result.

Stress-Free Detection Technologies for Pig Growth Based on Welfare Farming: A Review

Abstract. The detection of pig growth and monitoring of abnormal behaviors are key steps in pig breeding management. Using conventional methods to obtain information on growth and abnormal behavior causes stress to pigs, directly affects the number of live pigs for market, and decreases the quality of the pork. Moreover, this approach requires considerable labor, reduces economic returns, and does not meet the requirements of high-welfare farming. Compared to the conventional methods for obtaining growth parameters and data on abnormal behaviors, modern information technology provides a new method for stress-free growth detection and behavior monitoring in farmed pigs. This article first summarizes the importance of body size, body mass, and abnormal behaviors as well as the correlations among these factors. For the research on growth detection and behavior monitoring based on computer vision, radio frequency identification (RFID) and sensor technology, methods of detecting increases in body size and body mass and methods of monitoring abnormal behaviors are summarized separately. Through computer-computer vision technology, we found that the data sampling for growth and abnormal behaviors of the pigs was achieved without contact monitoring but, rather, occurred at the expense of complex data calculation and a higher illumination requirement during data collection. However, with the development of depth camera technology and improved product performance, technology based on high-precision depth cameras reduces the amount of data processing and complexity, making it possible to obtain real-time data on pig growth and abnormal behaviors. Moreover, with the advantages of no contact and no stress, the method conforms to the requirements of welfare farming. Keywords: Abnormal behaviors, Stress-free detection, Welfare farming.

High-resolution real-time 360° 3D model reconstruction of a handheld object with fringe projection profilometry.

The digitization of the complete shape of real objects has essential applications in fields of intelligent manufacturing, industrial detection, and reverse modeling. In order to build the full geometric models of rigid objects, the object must be moved relative to the measurement system (or the scanner must be moved relative to the object) to obtain and integrate views of the object from all sides, which not only complicates the system configuration but makes the whole process time-consuming. In this Letter, we present a high-resolution real-time 360° three-dimensional (3D) model reconstruction method that allows one to rotate an object manually and see a continuously updated 3D model during the scanning process. A multi-view fringe projection profilometry system acquires high-precision depth information about a handheld object from different perspectives and, meanwhile, the multiple views are aligned and merged together in real time. Our system employs stereo phase unwrapping and an adaptive depth constraint that can unwrap the phase of dense fringe images robustly without increasing the number of captured patterns. We then develop an efficient coarse-to-fine registration strategy to match the 3D surface segments rapidly. Our experiments demonstrate that our method can reconstruct the high-precision complete 3D model of complex objects under arbitrary rotation without any instrument assist and expensive pre/post-processing.

Design of a novel 3D tip-based nanofabrication system with high precision depth control capability

The design, analysis, and experimental investigation of a novel 3D tip-based nanofabrication system with high precision depth control capability is presented in this paper. Based on this system, a new depth control method, namely tip displacement-based closed-loop (DC) depth control methodology is proposed to improve the depth control capability. As the force-depth prediction with the commonly-used depth control method, i.e. the normal force-based closed-loop (FC) method, may depend on the machining speed, the machining direction, and the material properties, etc. Compared with the FC method, the DC method decreases the complexity and the high uncertainty. The tip feed system utilizes a non-contact force, i.e. the electromagnetic force, to adjust the tip displacement. Therefore, the tip support mechanism can be used to accomplish the tip-sample contact detection. Additionally, an active compensation method is proposed to eliminate the tilt angle between the sample surface and the horizontal plane. Otherwise the machining depth will change gradually, i.e. getting deeper or lower. Furthermore, a series of patterns have been fabricated on silicon sample surface with the proposed system and method. The maximum machining depth of a single scan reaches 300 nm, which is much larger than that of an atomic force microscope (AFM)-based nanofabrication system. The experimental results demonstrate that the system has advantages of distinguished depth control capability, high machining accuracy, and excellent repeatability, which diminishes the influence of above-mentioned factors on the machining depth. Also, the method has the potential of machining arbitrary 2D/3D patterns with well-controlled depth and high accuracy.

Computational Image Enhancement for Frequency Modulated Continuous Wave (FMCW) THz Image

In this paper, a novel method to enhance Frequency Modulated Continuous Wave (FMCW) THz imaging resolution beyond its diffraction limit is proposed. Our method comprises two stages. Firstly, we reconstruct the signal in depth-direction using a sinc-envelope, yielding a significant improvement in depth estimation and signal parameter extraction. The resulting high-precision depth estimate is used to deduce an accurate reflection intensity THz image. This image is fed in the second stage of our method to a 2D blind deconvolution procedure, adopted to enhance the lateral THz image resolution beyond the diffraction limit. Experimental data acquired with a FMCW system operating at 577 GHz with a bandwidth of 126 GHz shows that the proposed method enhances the lateral resolution by a factor of 2.29 to 346.2 μm with respect to the diffraction limit. The depth accuracy is 91 μm. Interestingly, the lateral resolution enhancement achieved with this blind deconvolution concept leads to better results in comparison with conventional gaussian deconvolution. Experimental data on a PCB resolution target is presented, in order to quantify the resolution enhancement and to compare the performance with established image enhancement approaches. The presented technique allows exposure of the interwoven fiber reinforced embedded structures of the PCB test sample.

Depth acquisition with the combination of structured light and deep learning stereo matching

The methods of structured light and deep learning are widely used in artificial vision to acquire a depth map of real-world scenes. In this paper, we propose a novel method of combining structured light and deep learning stereo matching to calculate the depth. To combat the problems with textureless areas of stereo matching, a pair of left and right side images with abundant structured light information is adopted to acquire a coarse depth map by unsupervised CNN networks. The coarse depth map is used for phase unwrapping, which is a bottleneck in the phase-based structured light method. Then, a fine and accurate depth map is obtained by phase matching. Compared with the traditional structured light method, the proposed method performs better on occluded and textureless areas. To evaluate the performance of our proposed method, an experimental platform is established and several experiments are conducted. Quantitative and qualitative experiments demonstrate that the proposed method can generate a high precision depth and relieve the occlusion in the structured light system.

Depth perception of moving objects viaing structured light sensor with unstructured grid

Abstract Structured light (SL) has been extensively researched and developed due to reconstruction the fast and highly accurate depth sensing. In order to achieve this goal, many approaches have been taken. In this article, we have taken some innovative work in the field for building a new type of fast, high-precision depth perception based on structured light. In addition, the basic ideas we propose are as follows: our model for slow motion dynamic scenes, the sparse depth maps are obtained based on the pair of unstructured rigid pattern with different color, and the high-quality depth images are optimized by the ridge line extraction. Finally, the calculated two depth maps based on the different color pattern merged, which to acquire the optimization depth. We further propose a model for the high-speed motion dynamic scene which apply a pair of projectors structure light. We show that, we actively use motion blur to estimate depth which is different from the previous approaches. The patterns are illuminated from two projectors and the motion blur of each line is accurately measured. The scene depth information is estimated by analyzing the length of the blur. As a result, we describe two experiments that we conducted on two different experimental equipment (ball and planner board). The first experiment was conducted with the planer board, measuring the slow moving object. In addition, the second experiment scene was conducted with the ball, which obtained the depth of the ultra-fast object. Our experimental results demonstrate that our approach achieves effective real-time measurement, can successfully obtain accurate depth information of moving objects.

High-Precision Depth Estimation Using Uncalibrated LiDAR and Stereo Fusion

We address the problem of 3D reconstruction from uncalibrated LiDAR point cloud and stereo images. Since the usage of each sensor alone for 3D reconstruction has weaknesses in terms of density and accuracy, we propose a deep sensor fusion framework for high-precision depth estimation. The proposed architecture consists of calibration network and depth fusion network, where both networks are designed considering the trade-off between accuracy and efficiency for mobile devices. The calibration network first corrects an initial extrinsic parameter to align the input sensor coordinate systems. The accuracy of calibration is markedly improved by formulating the calibration in the depth domain. In the depth fusion network, complementary characteristics of sparse LiDAR and dense stereo depth are then encoded in a boosting manner. Since training data for the LiDAR and stereo depth fusion are rather limited, we introduce a simple but effective approach to generate pseudo ground truth labels from the raw KITTI dataset. The experimental evaluation verifies that the proposed method outperforms current state-of-the-art methods on the KITTI benchmark. We also collect data using our proprietary multi-sensor acquisition platform and verify that the proposed method generalizes across different sensor settings and scenes.

Neuronal photoactivation through second-harmonic near-infrared absorption by gold nanoparticles

Optical activation of neurons requires genetic manipulation or the use of chemical photoactivators with undesirable side effects. As a solution to these disadvantages, here, we demonstrate optically evoked neuronal activity in mouse cortical neurons in acute slices and in vivo by nonlinear excitation of gold nanoparticles. In addition, we use this approach to stimulate individual epitheliomuscular cells and evoke body contractions in Hydra vulgaris. To achieve this, we use a low-power pulsed near-infrared excitation at the double-wavelength of the plasmon resonance of gold nanoparticles, which enables optical sectioning and allows for high spatial precision and large penetration depth. The effect is explained by second-harmonic Mie scattering, demonstrating light absorption by a second-order nonlinear process, which enables photothermal stimulation of the cells. Our approach also minimizes photodamage, demonstrating a major advancement towards precise and harmless photoactivation for neuroscience and human therapeutics.

Enhanced reference-plane-based dual-frequency absolute phase retrieval for depth measurement

Digital fringe projection is one of the most widely adopted optical non-contact methods used for high-precision depth measurement in industry. Nevertheless, new approaches with less measurement complexity and higher accuracy are still emerging in the research field. In this paper, we first analyze the efficiency of the reference-plane-based dual-frequency (RDF) pixel-by-pixel absolute phase retrieval method. The RDF method is able to provide a high-precision phase measurement, but the constraint of measurement depth range limits its application. We therefore propose an enhanced reference-plane-based dual-frequency (ERDF) pixel-by-pixel absolute phase retrieval method. The proposed ERDF method overcomes the constraint of measurement depth range and preserves the merits of the RDF method. In the ERDF method, the lower-frequency phase map is completely unwrapped with reference-plane-based single-frequency method, and the higher-frequency phase map is completely unwrapped using the lower-frequency unwrapped phase. The ERDF method is suitable for measuring objects with step-height surface discontinuities. Experiments have been conducted to demonstrate the advantages of the proposed ERDF method as compared to the RDF method.

"A review of bathymetric measurements from August 2018 campaign on the lower course of the Danube "

The presented paper shows a combination of different methods and techniques for high precision depth measurements. The results show the procedures for collecting field data from the bathymetric measurement campaign of August 2018, developed along the lower course of the Danube. Measurements were made over a total distance of 20 km. The main purpose of the measurements is to create bathymetric maps and to generate depth maps. To achieve the main purpose of this research paper, single beam bathymetric measurements were made, and they were combined with GPS RTK mode determination, in local Stereo 70 coordinate system. Repeated measurements in different quarters of the year will determine the morphometric changes in time of the riverbed on this study area. Furthermore, based on the bathymetric maps, in the future, a coherent river terrain model can be created, and it can be used for other purposes, like 1D/2D or 3D hydrodynamic modeling and flood inundation mapping.

DepthCut: improved depth edge estimation using multiple unreliable channels

In the context of scene understanding, a variety of methods exists to estimate different information channels from mono or stereo images, including disparity, depth, and normals. Although several advances have been reported in the recent years for these tasks, the estimated information is often imprecise particularly near depth discontinuities or creases. Studies have however shown that precisely such depth edges carry critical cues for the perception of shape, and play important roles in tasks like depth-based segmentation or foreground selection. Unfortunately, the currently extracted channels often carry conflicting signals, making it difficult for subsequent applications to effectively use them. In this paper, we focus on the problem of obtaining high-precision depth edges (i.e., depth contours and creases) by jointly analyzing such unreliable information channels. We propose DepthCut, a data-driven fusion of the channels using a convolutional neural network trained on a large dataset with known depth. The resulting depth edges can be used for segmentation, decomposing a scene into depth layers with relatively flat depth, or improving the accuracy of the depth estimate near depth edges by constraining its gradients to agree with these edges. Quantitatively, we compare against 18 variants of baselines and demonstrate that our depth edges result in an improved segmentation performance and an improved depth estimate near depth edges compared to data-agnostic channel fusion. Qualitatively, we demonstrate that the depth edges result in superior segmentation and depth orderings. (Code and datasets will be made available.)

A new navigation strategy for underwater robot in reactor pool combined propeller speed detection and dynamics analysis with sonar data correction

ABSTRACT The navigation reliability of remote-operated vehicle (ROV) in nuclear power plant is very important. It is necessary to carry out redundant navigation design using components with stronger anti-radiation performance compared with traditional navigation devices, to ensure relatively high precision navigation in the case of failure of traditional navigation equipment. The paper proposes a new navigation algorithm for ROV working in the reactor pool combined propellers’ rotating speed detection with hydrodynamics analysis and sonar data correction without using other sensors, just as a redundant navigation strategy. The designed ROV can work at any water depth thanks to the high-precision depth control algorithm. At first, the hydrodynamics analysis of ROV is completed. After that, the dead reckoning (DR) is implemented combined with rotating speed detection of high-performance propellers with dynamics analysis of ROV. Then, the error of DR is analyzed and compensated through repeated field experiments. At last, the sonar data are added to correct the deduced position of ROV when it stops. Field experiment verifies that the precision of the proposed algorithm is high enough to be used as a redundant navigation strategy for ROV working in the reactor pool of the nuclear power plant.

Signal restoration method for restraining the range walk error of Geiger-mode avalanche photodiode lidar in acquiring a merged three-dimensional image.

The fluctuation in the number of signal photoelectrons will cause a range walk error in a Geiger-mode avalanche photodiode (Gm-APD) lidar, which significantly depends on the target intensity. For a nanosecond-pulsed laser, the range walk error of traditional time-of-flight will cause deterioration. A new signal restoration method, based on the Poisson probability response model and the center-of-mass algorithm, is proposed to restrain the range walk error. We obtain a high-precision depth and intensity merged 3D image using this method. The range accuracy is 0.6cm, and the intensity error is less than 3%.

Restraint of range walk error in a Geiger-mode avalanche photodiode lidar to acquire high-precision depth and intensity information.

There exists a range walk error in a Geiger-mode avalanche photodiode (Gm-APD) lidar because of the fluctuation in the number of signal photoelectrons. To restrain this range walk error, we propose a new returning-wave signal processing technique based on the Poisson probability response model and the Gaussian functions fitting method. High-precision depth and intensity information of the target at the distance of 5 m is obtained by a Gm-APD lidar using a 6 ns wide pulsed laser. The experiment results show that the range and intensity precisions are 1.2 cm and 0.015 photoelectrons, respectively.

Diffusion-based three-dimensional reconstruction of complex surface using monocular vision.

Three-dimensional (3D) reconstruction based on optical diffusion has certain significant advantages, such as its capacity for high-precision depth estimation with a small lens, distant-object depth estimation, a monocular vision basis, and no required camera or scene adjustment. However, few mathematical models to relate the depth information acquired using this technique to the basic principles of intensity distribution during optical diffusion have been proposed. In this paper, the heat diffusion equation of physics is applied in order to construct a mathematical model of the intensity distribution during optical diffusion. Hence, a high-precision 3D reconstruction method with optical diffusion based on the heat diffusion equation is proposed. First, the heat diffusion equation is analyzed and an optical diffusion model is introduced to explain the basic principles of the diffusion imaging process. Second, the novel 3D reconstruction method based on global heat diffusion is proposed, which incorporates the relationship between the depth information and the degree of diffusion. Finally, a simulation involving synthetic images and an experiment using five playing cards are conducted, with the results confirming the effectiveness and feasibility of the proposed method.

Analytical model for CO(2) laser ablation of fused quartz.

This paper reports the development of an analytical model, with supporting experimental data, which quite accurately describes the key features of CO2 laser ablation of fused silica glass. The quantitative model of nonexplosive, evaporative material removal is shown to match the experimental data very well, to the extent that it can be used as a tool for ablative measurements of absorption coefficient and vaporization energy. The experimental results indicated that a minimum of 12 MJ kg-1 is required to fully vaporize fused quartz initially held at room temperature, which is in good agreement with the prediction of the model supplied with input data available in the literature. An optimal window for the machining of fused quartz was revealed in terms of pulse duration 20-80μs and CO2 laser wavelength optimized for maximum absorption coefficient. Material removal rates of 0.33μm per J cm-2 allow for a high-precision depth control with modest laser stability. The model may also be used as a parameter selection guide for CO2 laser ablation of fused silica or other materials of similar thermophysical properties.

Layered compression for high-precision depth data.

With the development of depth data acquisition technologies, access to high-precision depth with more than 8-b depths has become much easier and determining how to efficiently represent and compress high-precision depth is essential for practical depth storage and transmission systems. In this paper, we propose a layered high-precision depth compression framework based on an 8-b image/video encoder to achieve efficient compression with low complexity. Within this framework, considering the characteristics of the high-precision depth, a depth map is partitioned into two layers: 1) the most significant bits (MSBs) layer and 2) the least significant bits (LSBs) layer. The MSBs layer provides rough depth value distribution, while the LSBs layer records the details of the depth value variation. For the MSBs layer, an error-controllable pixel domain encoding scheme is proposed to exploit the data correlation of the general depth information with sharp edges and to guarantee the data format of LSBs layer is 8 b after taking the quantization error from MSBs layer. For the LSBs layer, standard 8-b image/video codec is leveraged to perform the compression. The experimental results demonstrate that the proposed coding scheme can achieve real-time depth compression with satisfactory reconstruction quality. Moreover, the compressed depth data generated from this scheme can achieve better performance in view synthesis and gesture recognition applications compared with the conventional coding schemes because of the error control algorithm.

Cenozoic sedimentary evolution of deepwater sags in the Pearl River Mouth Basin, northern South China Sea

Recent exploration revealed the high potential for hydrocarbon in the deepwater sags, Pearl River Mouth Basin, northern South China Sea. This paper reports its Cenozoic sedimentary evolution through backstripping of high precision depth data of interpreted sequence boundaries. Local backstripping parameters were mapped based on well and geophysical data. Sensitivity analysis indicates that the reliability of decompaction results were largely improved by using the local porosity parameters and the lithological parameters that vary with grid nodes. Maps of sedimentation rates of 17 sequences from 65 Ma to the present were constructed, showing the spatial–temporal variation of the sedimentation rate. Three rapid depositional stages, 65–32, 29–23.3, 18.5–10.5 Ma, and three slow depositional stages, 32–29, 23.3–18.5, 10.5–0 Ma, were identified with abrupt changes of sedimentary patterns. The three rapid depositional stages were in accord with syn-rifting stage, the first post-rifting depositional stage, and the second post-rifting depositional stage, respectively. And the three slow depositional stages were in keeping with three tectonic events respectively. Several significant sedimentary discontinuities at 32, 23.3 and 10.5 Ma were observed and discussed. The comparison between the study area and the ODP Site 1148 at 32–23.3 Ma indicates that before ~29 Ma the ODP Site 1148 was at similar sedimentation regime as that in the Baiyun and Liwan sags, but significant diversity appeared after ~29 Ma, when a large quantity of terrigenous sediments was trapped by strong post-rifting subsidence in the Baiyun and Liwan sags and could not reach the lower slope areas. Study revealed that the most rapid accumulation from 18.5 to 17.5 Ma might be mainly owing to the large sediment supply during this strong monsoon period.

J0202-3-3 水中ロボットの高精度深度制御手法の開発(生物の機能・運動と模倣(3))

J0202-3-3 水中ロボットの高精度深度制御手法の開発(生物の機能・運動と模倣(3))